mirror of
https://github.com/MariaDB/server.git
synced 2025-01-16 03:52:35 +01:00
d15c839c0d
This patch reduces the overhead of system calls prior to a query, for threadpool. Previously, 3 system calls were done 1. WSARecv() to get notification of input data from client, asynchronous equivalent of select() in one-thread-per-connection 2. recv(4 bytes) - reading packet header length 3. recv(packet payload) Now there will be usually, just WSARecv(), which pre-reads user data into a buffer, so we spared 2 syscalls Profiler shows the most expensive call WSARecv(16%CPU) becomes 4% CPU, after the patch, benchmark results (network heavy ones like point-select) improve by ~20% The buffer management was rather carefully done to keep buffers together, as Windows would keeps the pages pinned in memory for the duration of async calls. At most 1MB memory is used for the buffers, and overhead per-connection is only 256 bytes, which should cover most of the uses. SSL does not yet use the optmization, so far it does not properly use VIO for reads and writes. Neither one-thread-per-connection would get any benefit, but that should be fine, it is not even default on Windows.
80 lines
2.2 KiB
C
80 lines
2.2 KiB
C
/* Copyright (C) 2020 Monty Program Ab
|
|
|
|
This program is free software; you can redistribute it and/or modify
|
|
it under the terms of the GNU General Public License as published by
|
|
the Free Software Foundation; version 2 of the License.
|
|
|
|
This program is distributed in the hope that it will be useful,
|
|
but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
GNU General Public License for more details.
|
|
|
|
You should have received a copy of the GNU General Public License
|
|
along with this program; if not, write to the Free Software
|
|
Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1335 USA
|
|
*/
|
|
#pragma once
|
|
|
|
#include <WinSock2.h>
|
|
#include <windows.h>
|
|
|
|
struct st_vio;
|
|
|
|
struct win_aiosocket
|
|
{
|
|
/** OVERLAPPED is needed by all Windows AIO*/
|
|
OVERLAPPED m_overlapped{};
|
|
/** Handle to pipe, or socket */
|
|
HANDLE m_handle{};
|
|
/** Whether the m_handle refers to pipe*/
|
|
bool m_is_pipe{};
|
|
|
|
/* Read buffer handling */
|
|
|
|
/** Pointer to buffer of size READ_BUFSIZ. Can be NULL.*/
|
|
char *m_buf_ptr{};
|
|
/** Offset to current buffer position*/
|
|
size_t m_buf_off{};
|
|
/** Size of valid data in the buffer*/
|
|
size_t m_buf_datalen{};
|
|
|
|
/* Vio handling */
|
|
/** Pointer to original vio->vio_read/vio->has_data function */
|
|
size_t (*m_orig_vio_read)(st_vio *, unsigned char *, size_t){};
|
|
char (*m_orig_vio_has_data)(st_vio *){};
|
|
|
|
|
|
|
|
/**
|
|
Begins asynchronnous reading from socket/pipe.
|
|
On IO completion, pre-read some bytes into internal buffer
|
|
*/
|
|
DWORD begin_read();
|
|
|
|
/**
|
|
Update number of bytes returned, and IO error status
|
|
|
|
Should be called right after IO is completed
|
|
GetQueuedCompletionStatus() , or threadpool IO completion
|
|
callback would return nbytes and the error.
|
|
|
|
Sets the valid data length in the read buffer.
|
|
*/
|
|
void end_read(ULONG nbytes, DWORD err);
|
|
|
|
/**
|
|
Override VIO routines with ours, accounting for
|
|
one-shot buffering.
|
|
*/
|
|
void init(st_vio *vio);
|
|
|
|
/** Return number of unread bytes.*/
|
|
size_t buffer_remaining();
|
|
|
|
/* Frees the read buffer.*/
|
|
~win_aiosocket();
|
|
};
|
|
|
|
/* Functions related to IO buffers caches.*/
|
|
extern void init_win_aio_buffers(unsigned int n_buffers);
|
|
extern void destroy_win_aio_buffers();
|